To what extent do differences in cognitive tempo (reflection-impulsivity) identified in grade 4 account for variation in later academic achievement, independently of more general ability measures? Seventy children were administered the Matching Familiar Figures Test (MFF) in grade 4 and the Comprehensive Test of Basic Skills (CTBS) in grades 4, 5, and 6. Children identified as reflective based on performance on the MFF in grade 4 scored significantly higher on the CTBS achievement battery at all grade levels than those classified as impulsive. However, the two groups did not differ on achievement measures in grades 5 or 6 when scores were adjusted for initial differences in grades 4 CTBS. Although MFF error score and MFF response latency were each significantly predictive of grade 5 and grade 6 CTBS scores, neither of the MFF variables significantly improved the prediction of academic achievement when current level of achievement was controlled. Relations between reflection-impulsivity and academic achievement were stronger for middle SES children than for lower SES children.